Performance differences with gpt4
#4
by
Vezora
- opened
Unless open ai wildly messed up their benchmarks I think it’s pretty safe to assume they achieved better results with rlhf and the constant updates they do to their models. If you created a platform to gather rlhf from user’s using this model could achieve similar results to the latest gpt4 benchmark?
WizardLM
changed discussion status to
closed